Three long-term trials end with a quasi-equilibrium between soil C, N, and pH: an implication for C sequestration.
During the 1960s, wheat quotas were introduced in Australia to limit wheat production and this created interest in a broader range of cropping options such as lupins and rapeseed. Selective herbicides started to become available and direct drilling of crops became possible. In 1975, beef prices fell steeply and the superphosphate 'bounty' (a subsidy) was removed, increasing the cost of phosphorus (P) fertiliser to farmers and making pasture-based enterprises less attractive than cropping. These pressures forced an interest in alternative crops to wheat, in direct drilling, and in the possibility of continuous cropping. This spawned several long-term trials testing the stability of continuous cropping rotations involving grain legumes for nitrogen (N) supply, comparing frequent tillage with direct drilling, and comparing associated aspects of stubble management such as burning, surface mulching, or incorporation into soil. A primary concern in these trials was the management of N fertility in continuous cropping systems with no legume pasture phase (Reeves et al. 1984). The systems were to rely on N fixation and cycling from a grain legume phase, generally a lupin species, with little if any pasture phase. While the initial interest was in N fertility, data produced from the trials were used to subsequently identify acidification as a related degradation process (Helyar 1976). Similarly, data from some long-term trials, including one of the three reported here, are now being used to report the long-term change in soil carbon (C) concentrations under different land management practices (Chan et al. 2011). Most, if not all, of these long-term trials from the 1970s have now ceased, sometimes without adequate planning for their termination. This resulted in lost or incomplete datasets. We were in the fortunate position of being able to take soil samples from three of these long-term trials at about the time of their abandonment.
Here we report the final total %N, %C, and pH status of the main treatments in these long-term cropping trials. Our initial aim was to determine the long-term changes that took place in soil due to the various rotation, tillage, and stubble management practices that have become common across south-eastern Australia. The data lead to the hypothesis that there was a quantifiable relationship between soil %C, soil %N, and soil pH in these soils from the mixed farming zone, enabling us to determine the cost of C sequestration in terms of soil acidification.
Materials and methods
Three long-term trials in south-eastern Australia were found that included similar treatments and had similar design. All three trials had >26 years of recorded experimental data and all had been recently terminated due to a loss of funding. Thus, this work represents an opportunity to report the final effects on soil fertility of the farming practices investigated. Available data on the initial properties of the soils at the three sites is given in Table 1.
The first trial site was at Wagga Wagga (147[degrees]20'E, 35[degrees]05'S) on a red Kandosol (Isbell 1996) or Profondic Lixisol (WRB, McKenzie et al. 2004). Long-term annual average rainfall was 540 mm. The trial was known as 'Stable Agriculture Through Wheat And Grain Legumes' (SATWAGL). It consisted of 16 treatments initially, but only nine of those continued unaltered until the end of the trial. The treatments consisted of wheat-lupin (WL), wheat-wheat (WW), and wheat-clover (WC) rotations, with stubble burnt (B) or retained and mulched (R), and with either direct drilling (D) (knife points) or conventional (three-pass) tillage (P), the latter involving either off-set discs or scarifier depending on the stubble management (Table 2). More detail is given in Heenan et al. (1994); all wheat was Triticum aestivum and lupin was Lupinus angustifolius. The trial commenced in 1979 and ran until 2005 when a lack of funding caused its termination. The data we provided are from 2004, the last year for which soil analyses were conducted. Three replicates of each treatment (the wheat phase) were sampled. The soil %C and soil %N trends from 1979 to 2000 for this site were reported by Heenan et al. (2004), the soil C density data for this trial were reported by Chan et al. (2011), and the soil pH data to 1990 by Heenan and Taylor (1995). Yet, the relationship between these three variables in this trial had never been investigated.
The second and third trials were located near Rutherglen in north-eastern Victoria (146[degrees]30'E, 36[degrees]06'S) on a Yellow Sodosol (Isbell 1996) or Abruptic Luvisol (WRB, McKenzie et al. 2004). Long-term annual rainfall averaged 590 mm. The trial designated RGL6 was initiated in 1975 and consisted of phased wheat-lupin rotations (WL, LW), continuous wheat (WW), or near-continuous lupins (LL). Again, these were T. aestivum and L. angustifolius. In the first 6 years, these rotations were cultivated before sowing, but from 1982 onward they were direct-drilled. Stubble was burnt each year. Further detail is given in Reeves et al. (1984). The trial was abandoned after the 2008 season, after 34 years. We conducted our soil sampling in May (autumn) 2009. All four replicates were sampled.
The trial designated SR1 was in a paddock adjacent to RGL6. It was established in 1981 and consisted of crop rotations with tillage and stubble management contrasts, some of which changed over the years. In 1988, the rotations were interrupted by a crop of field peas to enable weed control. We report only those treatments that we know were maintained over the life of the trial and which are also the treatments most comparable with the other two trials reported here. These treatments were 2-year rotations of wheat-legume; the main difference between SRI and the other two trials is that the grain legume phase in the 2-year rotations alternated between lupin and faba bean (Vicia faba). Tillage and stubble treatments were conventional cultivation with stubble burnt (P-B), direct drilling with stubble burnt (D-B), and direct drilling with stubble retained (D-R). The trial was also abandoned after the 2008 season. We conducted our soil sampling in June (early winter) 2009. All eight replicates were sampled. Historic bulk density data were not available for the two Rutherglen sites, so C density calculations were not possible.
Soil sampling for all three trials was undertaken in autumn to early winter each year: 2004 for SATWAGL and 2009 for the two Rutherglen trials. At the SATWAGL site we collected five cores per plot at 0-5, 5-10, 10-15, and 15-20 cm depth, as had been the practice at that site since 1991. At the Rutherglen sites, knowing that the sites were being abandoned, we collected a larger mass of soil so that a wide range of analyses could be conducted. The larger soil mass was obtained by collecting 15 cores per plot for the 5-10 and 10-20 cm depths and 20 cores per plot for the lighter textured 0-5 cm depth. The extra cores at the surface 0-5 cm enabled us to collect enough soil mass for a range of biological measurements in addition to the chemical measurements.
Soil total %N was measured in a LECO furnace (LECO 1995). Soil pH was measured in 0.01 M Ca[Cl.sub.2] after end-over-end shaking for 1 h. Soil %C was measured by LECO for the Rutherglen samples. For the SATWAGL soils, %C was routinely measured by the Walkley-Black procedure (Walkley and Black 1934) but in 4 years the Walkley-Black method was compared with the LECO procedure, and hence a regression relationship used for conversion:
LECO %C = 1.074(Walkley-Black %C) + 0.112; n = 104, r = 0.975, P < 0.001
For comparison of treatments within the SATWAGL site we used the original Walkley-Black data. Comparison of %C and C: N across trial sites is based on the LECO conversion of the original Walkley-Black data for SATWAGL.
Analyses of variance, linear regressions, and multiple linear regressions were conducted in SigmaStat 3.1. Comparison with the initial mean data for any site is based on the error associated with the replicates for the site mean at the final sampling.
Table 3 shows the 2004 data for each treatment at each depth. As expected the highest %N occurred under WC rotations (treatments 11-13) and this was significant to at least 10cm depth. The direct-drilled treatments showed higher %N than conventionally cultivated treatments in the top 5 cm of soil, with little difference between treatments at 5-10 cm depth. The initial soil had 0.13%N in the 0-10cm layer. The three WC rotations exceeded this value after 25 years (T11, 0.16%N; T12, 0.21%N; T13, 0.18%N). The WL and WW rotations had fallen below 0.13%N, although T1 (WL DR) remained close at 0.119% N. The biggest loss was in WW without N fertiliser (T9), which decreased to 0.085%N. These observations are consistent with the change in soil %N in the 0-10cm layer for the period 1979-2000 reported by Heenan et al. (2004). They found no change in soil %N during time for treatments 1, 11, 12, and 13, and net losses for treatments 3, 4, 6, 9, and 10. Treatment 12 showed a positive trend but it was not significant at that stage.
The initial soil had 0.10%N in the top 10 cm. Table 4 shows the treatment means at each depth in 2009. Differences were limited to the top 10 cm, and the highest values were in the LW rotation. The means for these four treatments in the surface 0-10cm in 2009 were WW 0.092%N, LL 0.108%N, WL 0.095%N, and LW 0.122%N, suggesting that the only net increase in %N over 34 seasons occurred in the LW rotation, possibly due to recent crop residues in that rotation, as %N in WL remained unchanged.
Table 5 shows the treatment means at each depth. The only significant difference was at 0-5 cm, where direct drilling maintained slightly higher %N than conventional cultivation after 28 years where stubble was burnt. No initial %N data were available for the site.
In 2004 there were strong differences between treatments in the top 10cm of soil (Table 3). The WC rotations contained the highest %C at 0-5 cm depth, while the lowest values were associated with conventional cultivation, regardless of rotation. At 5-10 cm, the differences diminished but the WC rotations generally maintained higher %C while the rotation with no N input (T9) had the lowest %C. At 15-20cm, T13 (WC, PR) contained a higher %C than T6 (WL, PB). The initial soil had 1.3%C in the top 10 cm. This was maintained in T1 (1.27%) and T4 (1.22%) based on the error for the 2004 data. The %C decreased under T3 (1.15%), T6 (1.03%), T9 (0.95%), and T10 (1.04%) and increased under T11 (1.51%), T12 (1.88%), and T13 (1.65%). These observations are broadly consistent with the change in soil %C for the period 1979-2000 reported by Heenan et al. (2004). However, they found no change in soil %C for treatments 11 and 13, whereas 4 years further on we have found that these two treatments, and hence all three clover-wheat rotations (T11-T13, Table 2), have now resulted in increased %C relative to the 1979 site mean. Additionally, Heenan et al. (2004) reported net losses for treatments 3, 4, 6, 9, and 10, whereas we found that T4 was now within error of the initial site mean. The C density data were reported by Chan et al. (2011), as bulk density data were available for this trial.
As for the %N data, the only difference between treatments was the higher %C at 0-5 cm in the LW rotation (Table 4). The tendency was also evident at 5-10 cm. No initial %C data were available for the site.
As for the %N data, the tilled plots had lower %C than the direct-drilled plots but the effect was only significant at P = 0.06 despite the use of eight replicates (Table 5). The %C for all three treatments in 2009 was less than the initial site value of 1.9% but consistent with values of ~1.1% observed from 1983 onwards (data not shown).
Significant impacts of treatments on C:N ratio were only evident at 0-5 cm depth (Table 3). The ratio was highest in the treatment with no N input (T9) and narrowest on the three pasture rotations, particularly the direct-drilled treatment. The initial 10:1 ratio at 0-10 cm depth was generally maintained, the widest ratio being 11.31 (T9) and the narrowest 9.00 (T12).
Although the LW rotation had the highest %N and highest % C, the C: N ratio was highest in the surface 0-5 cm under WW (13.32) and lowest under LL (11.36), as might be expected from relative N inputs (Table 4). At 10-20 cm, the extremes were WL (12.68) and LL (9.92), the latter low ratio being consistent with the 0-5 cm layer.
A significantly lower C: N ratio occurred under direct drilling with stubble retained at 0-5 cm but the comparison did not quite reach significance at 5-10 cm (P=0.087, Table 5).
Trends in surface soil (0-10cm) pH at this site have previously been reported for the period 1979-1990 (Heenan and Taylor 1995), and profile acidification to 1991 was also reported (Conyers et al. 1996). The reported trends have continued (Table 3).
There were significant differences at all depths. At 0-5 cm, T10 had lower pH than all the other eight treatments. At 5-10cm, T9 and T10 (WW minus and plus N fertiliser, respectively) represent the extremes in pH, with the high relative pH in T9 persisting at 10-20 cm, and the acidic trend remaining apparent, although not significant, for T10 at 10-15 cm. The initial site pH values were 4.93 and 5.06 for 0-10 and 10-20cm depths, respectively. In 2004, the surface 0-10 cm means ranged from 4.71 (T9) to 4.15 (T10), indicating net acidification for all treatments. The 10-20 cm means ranged from 5.02 (T9) to ~4.4 (T4, T1, T10), indicating acidification in all treatments except T9.
While there were no significant differences in pH in the 0-5 cm layer, differences between treatments were evident in lower layers. The highest pH at 5-10 and 10-20 cm was under WW and the lowest under LL (Table 4). The initial pH values at 0-10 and 10-20 cm were 6.03 and 6.19, indicating a high degree of acidification at both depths and under all four treatments over 34 years.
Initial pH changes at the trial were reported by Coventry and Slattery (1991) and we report the continuation of those trends over the last 20 years in Fig. 1a. In general the steep initial rate of acidification has declined; however, there are exceptions. Under the LL rotation there has been no further acidification at 0-10 cm but a further 0.2-0.4 pH unit decrease at 10-20 cm. Under the WW rotation, acidification has continued steeply by 0.6 pH units at 0-10 cm depth but has slowed to 0-0.3 pH units at 1 0-20 cm depth. Under the WL and LW rotations, acidification at 0-10 cm has slowed over the last 20 years, with about a 0.2 pH unit decrease, but at 10-20cm the decrease was 0.34.7 pH units. Clearly acidification remains an issue for agricultural production in these soils.
As for RGL6, the treatment effects on pH were confined to 5-20cm depth (Table 5). The highest pH was associated with cultivation and with burning stubble; that is, the practice of direct drilling appears to have contributed to subsurface soil acidification. Figure 1b shows available data since 1982.
Relationships between %C, %N, and soil pH
These three long-term trials showed that the values of %C, %N, and pH were interconnected. The accumulation of C depended on the availability of N, or vice versa, since the C and N values were highly positively correlated (Fig. 2) and the C:N ratio covered a narrow range (Table 6). This close association applies to the shallow depth increments within the SATWAGL trial (Table 6).
There were no significant relationships between pH and %C or %N at the 0-5 cm depth where fresh surface residues and high root density interact with soil. Removal of the data for the 0-5 cm layer revealed a negative relationship between both %N and %C with the pH of soil at 5-20 cm depth. The trend, using treatment means, was similar for all three trials, and the two adjacent trials at Rutherglen have been combined.
SATWAGL: pH = 5.68 - 25.68N - 113.70[N.sup.2]; r = 0.671, n = 27, P < 0.01 (1)
pH = 5.58 - 2.51C + 1.19[C.sup.2]; r = 0.712, n = 27, P < 0.01 (2)
RGL6 + SR1: pH = 6.61 - 31.56N - 125.49[N.sup.2]; r = 0.771, n = 14, P < 0.01 (3)
pH = 5.91 - 2.57C + 0.76[C.sup.2]; r = 0.780, n= 14, P < 0.01 (4)
The increase in acidity with increase in %N (and %C) is steep at low concentrations of organic matter (e.g. from -0.4 to -0.7 pH units as %N rises from 0.04 to 0.1% (%C rises from 0.5 to 1.0%) but levels out (an extrapolation for the Rutherglen data) as %N > 0.1% (%C > 1%) and pH reaches 4.0 [+ or -] 0.2 (Fig. 3).
The 3-dimensional relationship between pH, %C, and %N is shown for the 5-20 cm depths in Fig. 4, viewed from two angles. Each site x depth combination is shown as a separate surface. The upward (grey or unfilled mesh) and downward (green or fine grid) protruding surfaces in the lower %C and %N range are due to the rotations involving lupin in RGL6. Overall, these surfaces stack to create the relationship described by the following multiple linear regression for %C:
%C = 0.531 + (8.973 * %N) - (0.101 *pH); n = 159, r = 0.949 (P < 0.001 for all three parameters) (5)
For any given site x depth combination, the trend is for the surface to move diagonally towards higher %C and %N simultaneously as pH falls. Notwithstanding the general negative relationship between %C and pH, at the higher %N end of the data range, it appears that %C maximises in the range 4.5 < pH < 5.0. Hence, while the relationship between %C and % N remains basically curvilinear over the whole surface (Fig. 4b), the relationship between %C and pH appears to have an optimum (a fold) in the upper %N range (Fig. 4a).
Most, if not all, of the long-term trials from the 1970s have now ceased due to budget reductions. Completion of long-term trials due to termination of funding often makes final soil sampling difficult or impossible. The work presented here represents a unique opportunity to report the concluding findings of three long-term trials investigating the effects on soil fertility of common farming practices implemented in southeastern Australia.
Some treatments are comparable between pairs of trials or across all three trials. The D-R treatment of SR1 compares with T1 of SATWAGL, differing in that the legume crop in SR1 alternates between lupin and faba bean. Both have mid-range pH for their respective trials, and mid-range %C and %N at least for the 0-5 cm depth.
The D-B treatment of SR1 is comparable with T4 of SATWAGL and with the WL and LW rotations of RGL6. All three treatments are in the mid-lower range of pH for their respective trials. There is no obvious pattern for either the %N or %C data.
The P-B of SR1 compares with T6 of SATWAGL. Both have low %N and low %C for their respective trials. Hence, long-term cultivation and stubble burning under continuous cropping resulted in a degraded soil at both sites.
The continuous wheat of RGL6 is comparable with T9 in SATWAGL, although the tillage differs. Both were the least acidified rotations of their respective trials. The SATWAGL T9 resulted in low relative %N, whereas for RGL6 this was only true at 0-5 cm, with the tendency apparent to 10 cm. The same was true for %C status. Both treatments had the widest C:N ratios at 0-5 cm (13.20 for LECO-corrected SATWAGL and 13.32 for RGL6). The association of low %N and higher relative pH for these two rotations without input of legume or fertiliser N indicates the importance of the N cycle to soil acidification.
Relationships between %C, %N, and soil pH
The comparisons of similar treatments between sites were much as expected. Further, the C:N range is consistent with the review of Kirkby et al. (2011). The contribution of C cycle process to soil acidification in agricultural systems has also been reported (Ritchie and Dolling 1985; Ridley et al. 1990; Lockwood et al. 2003). The influence of N fertility on decreasing soil pH is well understood (Helyar 1976). Williams and Donald (1957) observed relationships between C, N, and pH from their survey data in permanent pastures of the Southern Tablelands of New South Wales. The build-up of soil C over time was associated with an increase in soil cation exchange capacity (CEC) and a decline in pH, presumed to be due to the increased density of carboxylic and other acidic functional groups in soil organic matter. The three long-term trials utilised here have shown for the first time that the values of %C, %N, and pH are strongly interconnected within a range of tillage and stubble management practices common in the cropping systems of south-eastern Australia. We propose a conceptual biochemical model:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
When the rate of biomass input exceeds the rate of microbial respiration of soil organic matter (SOM), then there will be an increase in the SOM pool. As the C input humifies, soil effective CEC (ECEC) (due to RCO[O.sup.-]) will increase and pH will decrease. When microbial respiration exceeds the rate of biomass addition, the SOM pool will decrease in size. The ECEC will fall and pH will rise as the equation moves to the right with oxidation of RCO[O.sup.-]. Hence, there is a negative relationship between organic C (as humus, or non-particulates) and pH in soil. The relationship cannot be as precise as in chemical reactions, as there are other pH-changing processes in soil, such as net anion uptake (with product removal) and nitrification. Further, these other acidifying processes can also have indirect impacts on biomass production and, hence, C return to the soil. For example, the relationship between %C and pH appears to have an optimum (a fold) in the upper %N range (Fig. 4a). Possibly in this upper %N range, there is substantial acidity generated by nitrification and nitrate leaching; the fold possibly reflects the impact of aluminium toxicity on root growth at pH < 4.5 (Liu et al. 2004) and higher rates of microbial growth and activity on C metabolism as pH rises to > 5 (5.5-6.0 in water) (Aciego Pietri and Brookes 2008).
However, the relationship between %N, %C, and pH did not exist in the surface 5 cm of soil. The lack of such a relationship in the surface 5 cm is likely to be due to the dynamic nature of the forms and sizes of organic matter present, its decomposition, related pH change, and pH buffering associated with organic components. Below the surface 5 cm, the concentration of particulates present is decreased in the absence of surface residue return, allowing the relationships to become apparent. Although the surface 0-5 cm layer of the soil typically has a higher soil %C (Tables 3 and 5) than lower layers, the greater volume of soil in lower layers means that substantial quantities of C are present.
From a soil C sequestration perspective, the build-up of C comes at the cost of a required accumulation of soil %N; that is, the N is not being acquired by plants and is hence not contributing to production. Second, the build-up of %C and %N fertility comes at the cost of soil acidity. Rotation, tillage, and stubble practices combine to alter the quantity, quality (C:N), and the depth distribution of organic matter in a soil, but the three soil properties reported here seem to change in synchrony. There is quasi-equilibrium between these properties, with possible implications for the biochemical functioning of soil and an implication for C sequestration.
The implication for C sequestration is that if the desired build-up of %C in soil is associated with soil acidification, then the maintenance of agricultural production will require liming. Limestone releases C when it reacts with soil acids, plus liming can increase the respiration rate of native soil C. The C cost of quarrying, milling, transport, and spreading of limestone, in addition to these two well-known sources of C emission from liming soil, might therefore negate at least some portion of the gains from C sequestration as organic matter in soil.
A simple C budget is provided in Table 7 to estimate the magnitude of the liming cost associated with C sequestration. On the basis of Eqns 2 and 4, a pH decrease from 5.0 to 4.5 is associated with a 0.29% increase in %C at the Rutherglen sites and a 0.34% increase in %C at the SATWAGL site. On average, therefore, we can equate a 0.3% increase in %C with this 0.5 pH unit decrease from 5.0 to 4.5, at which point liming would be considered in order to maintain agricultural production. At a bulk density of 1.3 t [m.sup.-3], the 0.3% increase in %C equates to 3900 kg C [ha.sup.-1] 10 [cm.sup.-1]. The field pH buffering capacity of these soils, based on liming trials, is 0.4-0.5 pH unit [t.sup.-1] [ha.sup.-1] 10 [cm.sup.-1] of limestone. As each tonne of limestone reacts with soil acids, it will release 120kg C [ha.sup.-1] (limestone being 12%C). It will also mineralise 1-5% of the native C in SOM, which we have conservatively set at 1%C (13 t C [ha.sup.-1] 10 [cm.sup.-1]) for this calculated example, based on typical mixed farming soils as represented by these three trials. The C cost of mining, milling, transporting, and spreading limestone is derived from the study of Brock et al. (2012). It assumes a transport distance of 100 km. In Table 7 this has been extended to give a range of values based on the almost 300km transport cost for liming commonly required in the mixed farming zone of south-eastern Australia. Overall, we estimate in Table 7 that ~20-30% of the increase in soil %C is negated by the need for liming to maintain agricultural productivity. This might appear to be a small relative cost; however, the slow gains in soil C highlighted by Chan et al. (2011), of <700kg C [ha.sup.-1] [year.sup.-1], are slowed further by this additional cost.
The long-term effect of common cropping practices on soil %N, %C, and soil pH have now been reported following the termination of three long-term trials. Similarities between regional localities were apparent. Cropping systems that have no legume content and hence rely on fertiliser alone for N fertility are considered degrading to the soil resource in terms of acidification. The system maintaining the highest pH did so at the expense of running down soil %C and %N.
Based on long-term trial data, a quasi-equilibrium relationship between %N, %C, and soil pH has been identified. In order to sequester soil C, it will be necessary to increase N fertility. The gains in N and C in the system will induce a decrease in soil pH that will require liming to avoid soil degradation deleterious to agricultural production. The use of lime to rectify this degradation will cause C emissions that negate a proportion of the gains in C sequestered. Therefore, data obtained from long-term trials have identified that there is an environmental and C cost associated with C sequestration. This is in addition to any economic costs associated with fertiliser and lime inputs required to obtain higher soil C%. This work also highlights the value in long-term trials to provide useful information in contexts that were not conceived at the start of the trials.
We thank the Australian Research Council for funding the work. Karen Tymms, Richard Lowrie, and Ekaterina Simova-Samuelian assisted with aspects of the work.
Received 28 July 2011, accepted 3 October 2012, published online 13 November 2012
Aciego Pietri JC, Brookes PC (2008) Relationship between soil pH and microbial properties in a UK amble soil. Soil Biology & Biochemistry 40, 185-1861. doi: 10.1016/j.soilbio.2008.03.020
Brock P, Madden P, Schwenke G, Herridge D (2012) Greenhouse gas emissions profile for 1 tonne of wheat produced in Central Zone (East) New South Wales: a life cycle assessment approach. Crop & Pasture Science 63, 319-329. doi:10.1071/CP11191
Chan KY, Conyers MK, Li GD, Helyar KR, Oates A, Poile GJ, Barchia IM (2011) Soil carbon dynamics under different cropping and pasture management in temperate Australia: Results of three long term experiments. Soil Research 49, 320-328. doi: 10.1071/SR10185
Conyers MK, Heenan DP, Poile GJ, Cullis BR, Helyar KR (1996) Influence of dryland agricultural management practices on the acidification of a soil profile. Soil & Tillage Research 37, 127-141. doi:10.1016/0167-1987 (95)01005-X
Coventry DR, Slattery WJ (1991) Acidification of soil associated with lupins and grown in a crop rotation in north-eastern Victoria. Australian Journal of Agricultural Research 42, 391-397. doi: 10.1071/AR9910391
Heenan DP, Taylor AC (1995) Soil pH decline in relation to rotation, tillage, stubble retention and nitrogen fertilizer in S.E. Australia. Soil Use and Management 11, 4-9. doi: 10.1111/j. 1475-2743.1995.tb00487.x
Heenan DP, Taylor AC, Cullis BR, Lill WJ (1994) Long term effects of rotation, tillage and stubble management on wheat production in southern NSW. Australian Journal of Agricultural Research 45, 93-117. doi:10.1071/AR9940093
Heenan DP, Chan KY, Knight PG (2004) Long-term impact of rotation, tillage and stubble management on the loss of soil organic carbon and nitrogen from a Chromic Luvisol. Soil & Tillage Research 76, 59-68. doi:10.1016/j.still.2003.08.005
Helyar KR (1976) Nitrogen cycling and soil acidification. Journal of the Australian Institute of Agricultural Science 42, 217-221.
Isbell RF (1996) 'The Australian Soil Classification.' (CSIRO Publishing: Melbourne)
Kirkby CA, Kirkegaard JA, Richardson AE, Wade LJ, Blanchard C, Batten G (2011) Stable soil organic matter: A comparison of C:N:P:S ratios in Australian and other world soils. Geoderma 163, 197-208. doi: 10.1016/ j.geoderma.2011.04.010
LECO (1995) Instrumentation for characterisation of organic/inorganic materials and microstructural analysis. Instruction Manual FP-2000 Protein/Nitrogen Analyses, Form No. 200-558 May 1995. LECO Corporation, St Joseph, MI.
Liu DL, Helyar KR, Conyers MK, Fisher R, Poile GJ (2004) Response of wheat, triticale and barley to lime application in semi-arid soils. Field Crops Research 90, 287-301. doi: 10.1016/j.fcr.2004.03.008
Lockwood PV, Wilson BR, Daniel H, Jones MJ (2003) Soil acidification and natural resource management: Directions for the future. NSW Agriculture, Orange.
McKenzie NJ, Jacquier D, Isbell R, Brown K (2004) 'Australian soils and landscapes.' (CSIRO Publishing: Melbourne)
Reeves TG, Ellington A, Brooke HD (1984) Effects of lupin-wheat rotations on soil fertility, crop disease and crop yields. Australian Journal of Experimental Agriculture and Animal Husbandry 24, 595-600. doi: 10.1071/EA9840595
Ridley AM, Slattery W J, Helyar KR, Cowling A (1990) The importance of the carbon cycle to acidification of a grazed annual pasture. Australian Journal of Experimental Agriculture 30, 529-537. doi:10.1071/ EA9900529
Ritchie GSP, Dolling PJ (1985) The role of organic matter in soil acidification. Australian Journal of Soil Research 23, 569-576. doi: 10.1071/SR9850569
Walkley A, Black IA (1934) An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil Science 37, 29-38. doi:10.1097/ 00010694-193401000-00003
Williams CH, Donald CM (1957) Changes in organic matter and pH in a podzolic soil as influenced by subterranean clover and superphosphate. Australian Journal of Agricultural Research 8, 179-189. doi: 10.1071/ AR9570179
Mark Conyers (A,E), Philip Newton (B), Jason Condon (A), Graeme Poile (A), Pauline Mele (C,D), and Gavin Ash (A)
(A) EH Graham Centre for Agricultural Innovation, Department of Primary Industries NSW and Charles Sturt University, Wagga Wagga Agricultural Institute, Pine Gully Road, Wagga Wagga, NSW 2650, Australia.
(B) Formerly Victorian Department of Primary Industries, Chiltern Valley Road, Rutherglen, Vic. 3685; now PO Box 572, Wodonga, Vic. 3689, Australia.
(C) Department of Primary Industries, Biosciences Research Division, Victorian AgriBiosciences Centre, 1 Park Drive, Bundoora, Vic. 3083, Australia.
(D) School of Life Sciences, LaTrobe University, Bundoora, Vic. 3083, Australia.
(E) Corresponding author. Email: firstname.lastname@example.org
Table 1. Initial soil chemical properties of the three trials SATWAGL SRI RGL6 %C (A) (0-I0 cm) 1.3 1.9 -- %N (0-10cm) 0.13 -- 0.10 [pH.sub.ca] (0-10cm) 4.93 4.86 6.03 [pH.sub.ca] (10-20 cm) 5.06 -- 6.19 (A) Walkley-Black C. Table 2. Treatments at the SATWAGL site W, Wheat; L, lupins; C, subterranean clover Treatment Rotation Tillage Stubble Other TI WL Direct drill Retained T3 WL 3 x scarify Retained T4 WL Direct drill Burnt T6 WL 3 x scarify Burnt T9 WW 3 x scarify Burnt -N fertiliser T10 WW 3 x scarify Burnt +N fertiliser T11 WC 3 x scarify Retained Grazed (sheep) T12 WC Direct drill Retained Mown T13 WC 3 x scarify Retained Mown Table 3. Soil chemical properties at the SATWAGL site in 2004 after 25 seasons T, Treatment; W, wheat; L, lupin; C, subterranean clover; D, direct drill; P, three scarifyings; B, stubble burnt; R, stubble retained; N, N fertiliser. Within rows, means followed by the same letter are not significantly different at P=0.05; n.s., not significant Variable Depth T1 T3 T4 T6 T9 (cm) WL, DR WL, PR WL, DB WL, PB WW pH 0-5 4.51a 4.64a 4.74a 4.73a 4.78a 5-10 4.21b 4.32b 4.29b 4.38b 4.64a 10-15 4.24b 4.38b 4.24b 4.36b 4.82a 15-20 4.59b 4.75b 4.57b 4.74b 5.21a %C 0-5 1.52c 1.23d 1.48c 1.03d 1.02d 5-10 1.026c 1.07b 0.966c 1.02bc 0.88c 10-15 0.58 0.62 0.63 0.56 0.61 15-20 0.41ab 0.44ab 0.51 ab 0.36b 0.39ab %N 0-5 0.143d 0.106e 0.134d 0.097e 0.084e 5-10 0.094b 0.105ab 0.096b 0.097b 0.085b 10-15 0.057b 0.058b 0.065ab 0.057b 0.058b 15-20 0.048 0.051 0.051 0.043 0.047 C : N 0-5 10.7ab 11.6ab 11.1 ab 10.7ab 12.2a 5-10 11.0 10.2 10.0 10.6 10.4 10-15 10.2 10.8 9.8 9.8 10.5 15-20 9.0 8.9 10.1 8.5 8.4 Variable Depth TIO T11 T12 T13 P-value (cm) WW+N WC, PR WC, DR WC, PR pH 0-5 4.25b 4.54a 4.69a 4.65a <0.004 5-10 4.05e 4.27b 4.25b 4.29b <0.001 10-15 4.19b 4.42b 4.45b 4.41b <0.001 15-20 4.66b 4.76b 4.79b 4.75b 0.014 %C 0-5 1.12d 1.73bc 2.64a 1.98b <0.001 5-10 0.96bc 1.29a 1.12b 1.32a <0.001 10-15 0.57 0.68 0.66 0.72 n.s. 15-20 0.40ab 0.44ab 0.48ab 0.55a 0.025 %N 0-5 0.104e 0.184c 0.305a 0.216b <0.001 5-10 0.099b 0.137a 0.121a 0.139a <0.001 10-15 0.061b 0.074a 0.074a 0.075a 0.039 15-20 0.05 0.053 0.057 0.06 n.s. C : N 0-5 10.8ab 9.4bc 8.7c 9.2c 0.003 5-10 9.7 9.5 9.3 9.5 n.s. 10-15 9.3 9.2 9.0 9.7 n.s. 15-20 7.9 8.4 8.4 9.3 n.s. Table 4. Soil chemical properties at the RGL6 site in 2009 after 34 seasons Depth Variable (cm) WW LL WL LW P-value pH 0-5 4.62 4.60 4.68 4.75 n.s. 5-10 4.36a 4.04b 4.19ab 4.26ab < 0.001 10-20 5.25a 4.39e 4.81b 4.87b < 0.001 %C 0-5 1.52a 1.57a 1.57a 1.93b 0.03 5-10 0.80 0.86 0.76 0.99 0.09 10-20 0.49 0.45 0.54 0.63 n.s. %N 0-5 0.114a 0.138ab 0.125a 0.156b 0.019 5-10 0.069ab 0.078ab 0.065a 0.088b 0.041 10-20 0.046 0.045 0.042 0.056 n.s. C:N 0-5 13.3a 11.4c 12.7ab 12.3b < 0.001 5-10 11.7 11.1 11.6 11.3 n.s. 10-20 10.8ab 9.9b 12.7a 11.3ab 0.016 W, Wheat; L, lupins. Within rows, means followed by the same letter are not significantly different at P=0.05; n.s., not significant Table 5. Soil chemical properties at the SRI site in 2009 after 28 seasons Variable Depth P-B D-B D-R P-value (cm) pH 0-5 4.58 4.56 4.47 n.s. 5-10 4.48a 4.36b 4.35b 0.032 10-20 5.10a 4.79b 4.88ab 0.011 %C 0-5 1.09 1.24 1.12 0.06 5-10 0.77 0.76 0.74 n.s. 10-20 0.41 0.46 0.46 n.s. %N 0-5 0.093a 0.107b 0.103ab 0.038 5-10 0.069 0.071 0.072 n.s. 10-20 0.045 0.049 0.050 n.s. C:N 0-5 11.7a 11.6a 10.96 0.019 5-10 11.2 10.7 10.3 0.087 10-20 9.2 9.4 9.3 n.s. P, Three-pass conventional cultivation; D, direct drill; B, stubble burnt; R, stubble retained. Rotation was wheat followed by a grain legume (lupin or faba beans). Within rows, means followed by the same letter are not significantly different at P=0.05; n.s., not significant Table 6. Regression relationships between %N and %C at each site, and implied C : N ratio when forced through the origin Site Regression n r C:N SATWAGL N=0.088C (F) 36 0.994 11.4 N=0.098C - 0.014 36 0.964 RGL6 N=0.083C (F) 12 0.998 12.0 N=0.077C + 0.008 12 0.991 SRI N=0.091C (F) 9 0.997 11.0 N=0.075C + 0.014 9 0.995 SATWAGL 0-5 cm N=0.090C (F) 9 0.992 11.1 5-10 cm N=0.086C (F) 9 0.998 11.6 10-15 cm N=0.087C (F) 9 0.998 12.2 15-20 cm N=0.088C (F) 9 0.999 11.5 (F), Forced through the origin. All three depths are used for each site relationship and the individual depth relationships are also shown for the SATWAGL site Table 7. An estimated carbon budget for the limestone application required to counter 0.5 pH unit acidification associated with 0.3% soil C sequestration C gain C cost Factor (kg C [ha.sup-1] 10 [cm.sup-1] Soil organic matter 3900 accumulation of 0.3% Reaction of 1 t limestone 120 [ha.sup-1] Mineralisation of the 1 %C 130-650 native to the soil, at 1-5% of 13 t [ha.sup-1] Mining, milling, transport, 435-500 spreading of limestone (100-300 km cartage) Total 3900 685-1270 100% 18-33% Assumed bulk density of 1.3 3 g [cm.sup.-3]
|Printer friendly Cite/link Email Feedback|
|Author:||Conyers, Mark; Newton, Philip; Condon, Jason; Poile, Graeme; Mele, Pauline; Ash, Gavin|
|Date:||Oct 1, 2012|
|Previous Article:||Phosphorus loss and speciation in overland flow from a plantation horticulture catchment and in an adjoining waterway in Coastal Queensland,...|
|Next Article:||Prediction of the soil saturated paste extract salinity from extractable ions, cation exchange capacity, and anion exclusion.|